VMware Live Site Recovery (VLSR) VAMI login fails with "Unexpected status code: 503" due to a corrupted database
search cancel

VMware Live Site Recovery (VLSR) VAMI login fails with "Unexpected status code: 503" due to a corrupted database

book

Article ID: 443241

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

Symptoms:

  • Logging into the VLSR Virtual Appliance Management Interface (VAMI) fails with the error message: Unexpected status code: 503.

  • The /opt/vmware/support/logs/srm/vmware-dr.log file shows a complete halt in logging over an extended period (e.g., no entries recorded for several months).

  • The srm-server service is in an inactive or failed state.

    systemctl status srm-server
    . srm-server. service - VMware Site Recovery Manager Server
    Loaded: loaded (/usr/lib/systemd/ system/srm-server. service; disabled; vendor preset: enabled)
    Active: inactive (dead)

Environment

VMware Live Site Recovery 9.x

Cause

The VLSR VAMI login fails because the underlying srm-server service cannot start. The service repeatedly crashes due to severe database corruption within the Site Recovery Manager embedded database. Most of the required database schema and tables are missing entirely (with only the dr_product_info table remaining), causing critical SQL queries such as looking up the global_sequence relation to fail.

Cause Validation:

  • Attempt to restart the srm-server service fails. Reviewing the status reveals that srm-server service has crashed

    systemctl status srm-server
    srm-server. service - VMware Site Recovery Manager Server
    Loaded: loaded (/usr/lib/systemd/system/srm-server.service; disabled; vendor preset: enabled)
    Active: failed (Result: exit-code) since Fri 2026-05-29 08:44:16 UTC; 25min ago
    Process: 2774 Execstart=/opt/vmware/srm/bin/vmware-dr /opt/vmware/srm/conf/vmware-dr.xml (code=exited, status=240/LOGS_DIRECTORY)
    Main PID: 2774 (code=exited, status=240/LOGS_DIRECTORY)

    May 29 08:44:16 ########################### [1] : srm-server. service: Failed with result 'exit-code'.
    May 29 08:44:16 ########################### [1] : Failed to start Vware site Recovery Manager Server.
    May 29 08:44:16 ########################### [1]: srm-server.service: scheduled restart job, restart counter is at 3.
    May 29 08:44:16 ########################### [1] : Stopped VMware Site Recovery Manager Server.
    May 29 08:44:16 ########################### [1]: srm-server.service: start request repeated too quickly.
    May 29 08:44:16 ########################### [1] : srm-server. service: Failed with result 'exit-code'.
    May 29 08:44:16 ########################### [1]: Failed to start VMware Site Recovery Manager Server.

  • Reviewing the /opt/vmware/support/logs/srm/vmware-dr.log file immediately after an attempted service start shows that the database connection pool fails because essential relation entries do not exist:

    2026-05-29T08:44:15.310Z info vmware-dr [02774] [SRM@6876 sub=MainUtil] Initializing the DBManager
    2026-05-29T08:44:15.3102 info vmware-dr [02774] [SRM@6876 sub=Vdb] ConnectionPool: Creating 5 connections
    2026-05-29T08: 44:15.3312 warning vmware-dr [02774] [SRM@6876 sub=Vdb] Execute - SQL execution failed: alter sequence global_sequence increment by 1000
    2026-05-29T08:44:15.331Z warning vmware-dr [02774] [SRM@6876 sub=Vdb] Execute - Bind parameters:
    2026-05-29T08:44:15.331Z warning vmware-dr [02774] [SRM@6876 sub=Vdb] Execute - ODBC Error: "ODBC error: (42P01) - ERROR: relation "global_sequence" does not exist;
    -- > Error while executing the query" is returned when executing SQL statement "alter sequence global sequence increment by 1000"

  • Logging directly into the embedded database environment confirms that the database has been stripped of its tables, leaving it unrecoverable.

    Run the below commands to login to the DB

    /opt/vmware/vpostgres/current/bin/psql -d srmdb -U srmdb

    Type \d to list all the tables. In this case, only dr_product_info table exists.

    srmdb=> \d
              List of relations
    schema |      Name       | Type  | Owner
    -------+-----------------+-------+------
    public | dr_product_info | table | srmdb
    (1 row)

Resolution

Because the internal Site Recovery Manager database is completely corrupted and vital relational tables are missing, the appliance cannot be recovered or repaired.

To resolve this issue, you must redeploy the SRM appliance

Powered by Wolken Software